* Assertion failure in ISC BIND SIG query processing (CVE-2006-4095)
- Recursive servers
Queries for SIG records will trigger an assertion failure if more
than one RRset is returned. However exposure can be minimized by
restricting which sources can ask for recursion.
- Authoritative servers
If a nameserver is serving a RFC 2535 DNSSEC zone and is queried
for the SIG records where there are multiple RRsets, then the
named program will trigger an assertion failure when it tries
to construct the response.
* INSIST failure in ISC BIND recursive query handling code (CVE-2006-4096)
It is possible to trigger an INSIST failure by sending enough
recursive queries such that the response to the query arrives after
all the clients waiting for the response have left the recursion
queue. However exposure can be minimized by restricting which sources
can ask for recursion.
2006-04-09 Jarkko Hietaniemi <jhi@Chicxulub.local>
* Try to underline, highlight, and explain the fact that
String::Approx does not do a good job for comparing
strings "with fuzz", use the Levenshtein et al for that.
* aindex() might return "too early" indices if either the pattern
or the text contain repetitive characters, this seems to be a
tricky defect to fix and somewhat conflicting with our model
(return "as early as possible" matches, just like regular
expressions) (to get "as late as possible" matches one would
basically have to keep retrying at later indices until one
fails), so for now only document this known problem.
The same goes for aslice().
* [INTERNAL] use Test::More (not 100% yet)
* [INTERNAL] apse.c: do not reset text_position_range in apse_reset
(thanks to Helmut Jarausch)
* [INTERNAL] apse.c: add #include <assert.h>.
* [INTERNAL] apse.c: small signed/unsigned cleanups.
* Released as 3.26.
2005-05-24 Jarkko Hietaniemi <jhi@iki.fi>
* Pure documentation cleanup release to address
http://rt.cpan.org/NoAuth/Bug.html?id=12196
"Small String::Approx Pod issue"
* Released as 3.25.
2005-01-02 Jarkko Hietaniemi <jhi@iki.fi>
* Pure documentation cleanup release to address
http://rt.cpan.org/NoAuth/Bug.html?id=6668
"Unfortunate perldoc rendering of String::Approx"
* Released as 3.24.
- If SSL_cipher_list is not given it uses the openssl default
instead of setting it to 'ALL:!LOW:!EXP' like before. The old
value included ADH and this might be a bad idea, see BUGS why.
Resolves PR pkg/34392 by Martin Wilke
- Added some examples to the examples/ directory.
- Updated the MANIFEST.
- Added the :warndie tag and the internal Error::WarnDie package that
provides custom __WARN__ and __DIE__ handlers.
- Added the flush() method from Alasdair Allan.
- Fixed a test in t/05text-errors-with-file-handles.t to work on
MS Windows due to File::Spec and require inconsistency.
- Fixed https://rt.cpan.org/Ticket/Display.html?id=3291
- According to https://rt.cpan.org/Ticket/Display.html?id=6130 - made
the auto-conversion of textual errors to object customizable.
- Fixed the conversion of textual messages to Error::Simple when
they contain information about an open filehandle. (as reported in
http://rt.cpan.org/Ticket/Display.html?id=6130 )
- Added POD to the lib/Error/Simple.pm module.
- Added the lib/Error/Simple.pm module (that just "use"'s Error) so
one can say "use base 'Error::Simple';' Added an appropriate test.
Fixes: http://rt.cpan.org/Public/Bug/Display.html?id=17841
- Added Scalar::Util to the dependencies in Makefile.PL.
- Fixed http://rt.cpan.org/Public/Bug/Display.html?id=18024 and a related
exception thrown because ->isa was called on something that was not
certainly an object.
- Moved Error.pm inside the distribution to reside under lib/.
notable pkgsrc-relevant changes:
* Since support for non-system sound events has been removed from
GNOME 2.14, the mail-notification.soundlist file is no longer
installed
* desktop file forced to share/applications from etc/xdg/autostart
(see below)
Other changes:
[various translations updated]
* Evolution changes:
* Evolution crashed when a message without Subject or
From header was received: fixed
* Mail Notification could crash when removing an
Evolution mailbox: fixed
* Core changes:
* An alternate, compact mail summary has been added
* A "Display seen messages" option has been added
* The status icon can now be displayed permanently
* The "Delay between mail checks" option is now
mailbox-specific
* The properties dialog has been reorganized
* The main window has been removed
* The status icon is now activated with a single click
(http://bugzilla.gnome.org/show_bug.cgi?id=148829)
* The GTK+ bugs
(http://bugzilla.gnome.org/show_bug.cgi?id=321886)
which caused occasional crashes have been worked
around
* If the MAIL environment variable is not defined or
invalid, the system mailbox is now looked up in
/var/spool/mail/username and /var/mail/username
(based on a patch from Benoît Rouits)
* Since ~/.gnome2/mail-notification/mailboxes.xml may
contain sensitive information, it is now saved with
permissions set to 0600 (useful for systems where
home directories have lax permissions)
* The "cannot enable immediate notification" error
dialog has been removed
* The --list-features command line option has been
removed (feature information is now displayed in the
output of the --version option)
* The message handling performance has been improved
* If there are no mailboxes on startup, the properties
dialog is now displayed
* The "Start Mail Notification on GNOME login" option
no longer worked with GNOME >= 2.14 and has been
removed: a .desktop file is now installed in the xdg
autostart directory instead
* The mail reader setting has been removed; the mail
reader configured in GNOME (Desktop -> Preferences
-> Preferred Applications) will now be used
* The tooltip width is now limited to the screen width
* When coupled with a patched GNOME Panel (as shipped
by FreeBSD and Debian, for instance), the status
icon now supports panel transparency (patch
forwarded by Bernat Tallaferro)
* mbox, Maildir, POP3, IMAP, Evolution and Sylpheed changes:
* Messages are now classified as unseen and seen; seen
messages can be ignored by unchecking the "Display
seen mail" option
* mbox changes:
* Compressed mailboxes (gzip and bzip2) are now
handled transparently, and support immediate
notification
* POP3, IMAP and Gmail changes:
* A regression which allowed mail checks to overlap
has been fixed
* POP3 and IMAP changes:
* If the server advertises the STLS/STARTTLS
capability but replies ERR/BAD when the
STLS/STARTTLS command is issued, Mail Notification
now logs out instead of proceeding
* If multiple mailboxes have the same type, hostname,
port, connection type and username, Mail
Notification will now detect that the mailboxes
share the same account and will only prompt for the
password once per session (when the password is not
entered in the properties dialog, that is)
* POP3 changes:
* Non-UTF-8 server responses are now accepted
* IMAP changes:
* Non-ASCII server responses are now accepted (these
responses violate RFC 3501, but we follow the RFC
793 robustness principle "be liberal in what you
accept"); fixes connection problems with
internationalized Exchange servers, which violate
RFC 3501 by sending non-ASCII characters in their
greeting message
* Empty continuation responses are now allowed (fixes
authentication with some SASL mechanisms)
* Play nice with uncompliant servers by using
BODY.PEEK rather than BODY, so that they do not set
the \Seen flag (patch from Jacob Berkman)
* The IDLE mode is now cycled every 29 minutes, as
advised by RFC 2177
* Evolution changes:
* Support for the Evolution 2.4 and 2.6 branches has
been added; untested support for the Evolution
2.7/2.8 branch has been added; the Evolution source
tree is no longer needed for Evolution >= 2.4
* Sylpheed changes:
* With the help of a Sylpheed patch, Mail Notification
can now reliably avoid race conditions which could
occur when Sylpheed was running (see the
--enable-sylpheed-locking discussion in the INSTALL
file)
* New translations:
* Czech (Pav Lucistnik)
* Brazilian Portuguese (Claudio André)
* Dutch (Marcel J. Zwiebel)
* Japanese (Hiroshi Hasebe)
- ACPI support
- PC VGA BIOS fixes
- switch to OpenBios for SPARC targets (Blue Swirl)
- VNC server fixes
- MIPS FPU support (Marius Groeger)
- Solaris/SPARC host support (Ben Taylor)
- PPC breakpoints and single stepping (Jason Wessel)
- USB updates (Paul Brook)
- UDP/TCP/telnet character devices (Jason Wessel)
- Windows sparse file support (Frediano Ziglio)
- RTL8139 NIC TCP segmentation offloading (Igor Kovalenko)
- PCNET NIC support (Antony T Curtis)
- Support for variable frequency host CPUs
- Workaround for win32 SMP hosts
- Support for AMD Flash memories (Jocelyn Mayer)
- Audio capture to WAV files support (malc)
pkgsrc changes:
On systems with gcc4, force to use pkgsrc gcc34 to make it build
on NetBSD >=4.0.
At some point, pkgsrc started setting a different MAKE_ENV in the
install phase than in the build phase. This caused scons to report that
"the build action changed", which caused scons to rebuild csound5 and
install it with a broken configuration.
This hack saves MAKE_ENV to a file in WRKDIR, then uses that file to set
the environment in the install phase.
Thu Jul 27 04:53:01 CEST 2006 mikulas:
Fixed pagedown and search bugs in braille terminal introduced in
links-2.1pre20
Sun Jul 16 16:35:54 cet 2006 mikulas:
Link to "" redirects to current file rather than current directory
(fixes some forms)
Sun Jul 16 00:08:30 cet 2006 mikulas:
Better ftp directory parsing (process correctly file names starting with
year)
Sat Jul 15 16:37:36 MET 2006 user:
Bigger default textarea
Mon Jul 10 22:53:09 MET DST 2006 mikulas:
Use pkg-config to search for libpng and pcre
Tue Jun 27 23:50:43 cet 2006 mikulas:
Fixed display of multiple consecutive spaces when searching
Tue Jun 27 23:14:35 cet 2006 mikulas:
View as space when searching
Sun Jun 18 16:11:57 MET 2006 user:
Command "kill background connections" aborts also keepalive connections
Sat Jun 17 21:43:01 cet 2006 mikulas:
Removed SDL support until someone fixes it.
Remaining bugs:
palette not installed with 256-color svgalib support
drawing doesn't synchronize with softcursor (seen on svgalib)
stupid keyboard handling (SDL has the ability to query
operating system keymap and translate keys to unicode
but this driver didn't use it)
doesn't support virtual devices with Alt-1 ... Alt-0, like
svgalib and framebuffer
reportedly crashes with sdl-1.2.10
If you want SDL support continued, take file sdl.c from previous
version of links, fix the above issues and resubmit it.
Sat Jun 17 21:08:04 cet 2006 mikulas:
Fixed exponential memory consumption with table cell colspan/rowspan
Sat Jun 17 20:30:49 cet 2006 mikulas:
Do not try to interpret <script src=...> if server returns HTTP error
Sat Jun 10 23:52:00 cet 2006 mikulas:
Decode %AB sequences in position tags
Fri Jun 9 18:32:55 cet 2006 mikulas:
Accept broken meta refresh tags (fixes www.seznam.cz)
Fri Jun 9 02:14:42 MET 2006 user:
Shift-Q exits without question
Sat Jun 3 21:47:19 cet 2006 mikulas:
Fixed a bug that <body onload=...> was not processed when there were no
other scripts on that page
Sat Jun 3 19:31:35 CEST 2006 mikulas:
Fixed crash in javascript regular expressions
Wed May 31 18:25:32 CEST 2006 mikulas
Fixed mouse in Cygwin
Allowed to run in xterm or rxvt in Cygwin
Pkgsrc changes:
- Added patch-ac to change the number of skipped online tests (still
incorrect).
Changes since version 0.57:
===========================
Feature: hooks for DLV support in Net::DNS::SEC
added hooks for DLV support which is/will be available in
Net::DNS::SEC as of subversion version 592 (Tests are done agains
the subversion number, not against the perl release version)
Net::DNS::SEC version 0.15 will have DLV support.
Partly Fixed rt.cpan.org 18940
djhale noticed a number of error conditions under which de
udp_connection in Nameserver dies. We now print a warning instead
of dying.
Fix rt.cpan.org 18958
Fixed typebyname croak for SIGZERO. Acknowledgements to djhale.
Optimize rt.cpan.org 11931
Hanno Stock optimized the method to get the list of available
interfaces in Win32. I have only done very rudimentary tests on
my Windows XP system.
Fix dependency on "CC" rt.cpan.org 19352
The Makefile.PL depended on availabliltity of "cc" and would bail
out on systems where gcc is exclusivly available. Thanks to Rob
Windsor for noticing and patching.
Fix compressed dnames in NAPTR/SRV
Clayton O'Neill noted that the the domain names in in the NAPTR
and SRV RRs rdata were subject to name compression which is not
conform the specs. Also see RFC 2782 and 2915.
Fix rt.cpan.org 18897
Zero-length rdata in TXT fixed (Acknowledgements to Roy Arends)
Fix rt.cpan.org 18785
SPF would not work unless the TXT RR was not loaded first. No
wonder, SPF fully inherits TXT and loading of TXT.pm is therefore a
prerequisit.
Fix rt.cpan.org 18713
Net::DNS::Resolver now deals gracefully with persistent sockets
that got disconnected. It will first try to connect again to the
socket and if that fails it will try to connect to the next
available nameserver. tcp_timeout() is the parameter that
determines how long to wait during a reconnect.
Fix rt.cpan.org 18268
Added reference to RFC in croak message for label length > 63 in
dn_comp().
Fix rt.cpan.org 18113
The inet6 tests contained another bug when online-tests were disabled.
Klaus Heinz discovered and provided a patch.
PR pkg/34410.
Revision history for Perl extension XML::Atom::Stream:
0.09 Sun Aug 13 17:57:32 JST 2006
- Now when your callback function throws an exception, the
connection will be lost. This is now the official way to
disconnect from the Atom stream explicitly.
- Added POD document for METHODS
- Fixed code to get the number of youreTooSlow
- Added timeout option to new, which defaults to 15
0.08 Sat Aug 12 18:15:37 JST 2006
- Removed $DEBUG and make it $self->{debug}
- Added 'reconnect' option to automatically reconnect to the stream when disconnected
0.07 Fri Aug 11 00:46:24 JST 2006
- Wrap callback in eval again not to die
0.06 Fri Aug 11 00:17:36 JST 2006
- Don't die if youreTooSlow
0.05 Thu Aug 10 23:33:00 JST 2006
- Fixed $VERSION gotcha
- Added handling for <sorryTooSlow>
0.04 Thu Aug 10 22:58:07 JST 2006
- Update XML::SAX::ParserFactory ParserPackage to use XML::LibXML::SAX::Better
which is an extended SAX handler by Djabberd project. Now
this module actually works.
0.03 Thu Jun 15 08:23:42 JST 2006
- Support extra xmlns attributes declaration (Thanks to Byrne Reese)
- Don't die even if parsing fails
- Update eg/stream-client
PR pkg/34420.
Revision history for XML::Atom:
0.23 2006.08.27
* Fixed the method to get xml:lang and xml:base due to the
XML::LibXML 1.60 change which invalidated it.
0.22 2006.07.24
* Refactored internal element accessors by eating the new dog food
mk_elem_accessors and mk_object_list_accessor.
* Added support of Atom 1.0 <category> parse and generation
* Added $thing->links and $thing->categories as a moniker method
that returns an array reference in a scalar context
* Fixed a bug in $content->body() where it accidentally thinks the
content is not a valid Unicode string even if it is, if you call
eval {} in elsewhere in the code and $@ is left set.
(Thanks to Chris Dent for the patch)
0.21 2006.07.13
* propagate $entry's version when we create content element off of entry
using $entry->content("foo") syntax.
(Thanks to Simon Wistow for spotting this bug)
0.20 2006.07.12
* Lots of refactoring, which simplifies much of the code in the various
construct modules (Person, Link, etc), and which should also make it
much more straightforward to add extension classes in the future.
* Added global $XML::Atom::ForceUnicode flag to return everything as Unicode flagged
(Suggested by many people)
* Added global $XML::Atom::DefaultVersion flag to set default
version number for generated Atom feed. Defaults to 0.3 (for backward compatibility)
* Added support for atom:content @type in Atom 1.0 feeds
(Suggested by many people, especially Chris Dent and Andy Lester from Socialtext)
0.19 2006.03.19
* Fix 0.18 bug where renaming stuff was totally broken.
0.18 2006.03.16
* Support Atom 0.3 -> 1.0 renaming bits (issued -> published,
modified -> updated, tagline -> subtitle)
* $atom->content->body doesn't return Unicode flagged variable
anymore, even if it's text/ data. Now it just returns UTF-8 bytes.
(Thanks to Garth Webb)
PR pkg/34399.
Revision history for XML-Atom-SimpleFeed:
0.8 2006 Jun 3
- Multiple consecutive internal refactors; code structure is now
actually satisfactory
- Handles multiple authors and contributors
- Support for icon and logo elements
- Big POD cleanup
- Use Carp::Clean to get rid of silly $Carp::CarpLevel juggling
- ***BACKWARDS INCOMPATIBLE API CHANGE***:
Elements such as C<link> which may appear multiple times are no
longer specified in an anonymous array, but simply given
repeatedly.
- ***BACKWARDS INCOMPATIBLE API CHANGE***:
Atom 0.3 element and attribute names are no longer supported. (No
point keeping a lot of deprecation code around in the face of a
change like the above.)
- ***BACKWARDS INCOMPATIBLE API CHANGE***:
Suppress the default C<generator> element requires calling the
C<no_generator> method instead of passing a C<generator> key to
C<new> with an undefined value.
- ***BACKWARDS INCOMPATIBLE API CHANGE***:
Well, since I'm at it, the C<save_file> method is no longer
supported. C<print> now takes a handle, though.
- Cleaned up errors and warning messages and got rid of DIAGNOSTICS
section in POD
0.8_004 2006 May 10
- Brownbag upload: forgot to update ./Changes in 0.8_003
0.8_003 2006 May 10
- Minor incremental progress; various bugfixes, some refactor.
0.8_002 2006 Apr 9
- Use builtin XML writer instead of SAX for output. This
eliminates huge amounts of redundancy.
- Big improvements in the distribution of responsibilities for
deprecation and validation checks.
- Array-based implementation rather than inside-out objects.
- Internal structure is now more logical and consistent.
0.8_001 2005 Sep 28
- Emit Atom 1.0. Documentation updated to reflect Atom 1.0.
Usage according to Atom 0.3 will transparently generate 1.0
elements but emit deprectation warnings.
- Remove _generate_entry_id and use HTTP URLs as IDs by default.
Using tag: URIs is useful for generating the ID once, up
front, so that it won't change even if the permalink does --
if the ID is generated from the permalink, we might as well
use the permalink directly.
- Use XML::SAX::Writer instead of XML::Simple for output.
